Browse By Repository:


Develop fuzzy logic to monitor the health of transformer by using DGA

Muhammad Zulhilmi , Mohd Hanafi (2015) Develop fuzzy logic to monitor the health of transformer by using DGA. Project Report. UTeM. (Submitted)

[img] Text

Download (593kB)


The gases produced in oil filled insulation transformer can be used for detecting incipient faults. Dissolve gas analysis (DGA) is a famous approach method for the detection of incipient fault condition within the oil immersed transformers. The presence of key gasses is supervised and checked regularly. DGA is the most reliable and effective technique for condition monitoring. The aim of this project is to get the best method for detecting incipient fault of the actual transformer. There are number of method to detect the incipient fault such as: Key Gas, Roger Ratio, Doernenburg, IEC Ratio and Duval Triangle. As a result, various method techniques do not give the same result for the same oil sample. The method based more on personnel experience than mathematical formulation. For multiple fault condition due to different combination of produced gas, DGA method could not adapt to diagnose the problem. To overcome this problem, fuzzy logic approach is introduced to get an approximate to estimate ambiguity involved in a problem. Fuzzy logic approach help overcome the DGA limitation of multiple fault condition due to the mixing up of the produce gases. MATLAB program is developed to automatic simulate the method use, thus by comparing both the simulation fault and actual fault of the present transformer the best method of DGA is selected.

Item Type: Monograph (Project Report)
Uncontrolled Keywords: electric transformers, electric power transmission
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions: Library > Projek Sarjana Muda > FTK
Depositing User: Noor Rahman Jamiah Jalil
Date Deposited: 14 Dec 2015 06:07
Last Modified: 14 Dec 2015 06:07

Actions (login required)

View Item View Item


Downloads per month over past year